On-shell pair production neglected

Asked by Sokratis Trifinopoulos on 2020-01-30

Hi Madgraph team,

I am introducing a vector mediator V which couples to b quarks and tau neutrinos and also a new top-like, colored fermion psi and a right-handed neutrino-like fermion chi. The mass of V is 1 TeV, that of psi is 400 GeV and chi 300 GeV. The coupling of the interaction is 1.

I am generating the process:
p p > psi psi~, ( psi > b ta+ chi) , ( psi~ > b~ ta- chi~ )
trying to produce the psi psi~ on shell and then have them decay via off-shell V to the final products.

The cross-section for p p > psi psi~ is 7.7 pb, but when I launch the whole decay chain the cross-section becomes ~10^(-9). I would expect them to be equal, since the branching ratio is 1.

I do not understand this discrepancy.
Thanks in advance for your time.

Question information

Language:
English Edit question
Status:
Solved
For:
MadGraph5_aMC@NLO Edit question
Assignee:
No assignee Edit question
Solved by:
Olivier Mattelaer
Solved:
2020-02-04
Last query:
2020-02-04
Last reply:
2020-02-03

Olivier Mattelaer suggests this article as an answer to your question:
FAQ #2442: “why production and decay cross-section didn't agree.”.

Did you check that FAQ?
Likely the issue in your case is that you did not set the total width correctly.

Cheers,

Olivier

Thanks for the help!
You are correct, I had to calculate the width and put it manually in the param card. Previously, I was using auto thinking that it would take care of it, but apparently it doesn't work for very narrow width. Do you might know why?

Hi,

This is probably a bug in the auto-width.
I would need the model/benchmark to check what's going in this case.

Cheers,

Olivier

So, if you are interested in checking the bug, should I send you the .fr or ufo? (In that case how do I do it, by e-mail?)

Thanks Olivier Mattelaer, that solved my question.

You can either send it to me by email (<email address hidden>) with a link to this thread.
Of you can convert this tread to a bug report, and this will allow you to attach file.

Cheers,

Olivier

 Hi Olivier,
I am sorry for the delayed answer. Here is the model file and the UFO that we are using. The auto width does not work for the off-shell decay of the psi after it's pair-produced:Thanks in advance for your time!
Cheers,Sokratis On Tuesday, February 4, 2020, 09:08:25 PM GMT+1, Olivier Mattelaer <email address hidden> wrote:

 Your question #688353 on MadGraph5_aMC@NLO changed:
https://answers.launchpad.net/mg5amcnlo/+question/688353

Olivier Mattelaer posted a new comment:
You can either send it to me by email (<email address hidden>) with a link to this thread.
Of you can convert this tread to a bug report, and this will allow you to attach file.

Cheers,

Olivier

--
You received this question notification because you asked the question.

Hi,

attachment are not allowed for question.
you can either send me the email directly to me (not reply by email to launchpad) or convert this to a bug report

Cheers,

Olivier

Hi,

Ok, this is not a bug actually it is done by design.
When computing the width with "compute_width psi"
you will see this warning:
partial width of particle psi lower than QCD scale:3.5562e-06. Set it to zero. ((-15, 5, 9000005))

We do not trust decay when the scale is lower than the QCD. The True width is in this case unknown to my point of view.
And therefore consider such partial width to zero.

Cheers,

Olivier

Hi,

so this essentially means that the colored particle \psi hadronizes and this is the reason that you don't trust the decay rate.
If this is the case, it is actually very interesting, thank you for noticing it!

Hi again,

sorry for getting back to it, but it would be very interesting to know how Madgraph computes this lower bound for the width. Is it some kind of perturbativity bound?
Thanks again for the quick responses.

This is an hardcoded value.

Cheers,

Olivier

So it is always 3.5562e-06? In that case why is it 10^3 smaller than Lambda_QCD = 218MeV ?

Hi,

The 3.5562e-06 printed above is the computed partial width.
That value is computed according to this paper: arXiv:1402.1178 <http://arxiv.org/abs/arXiv:1402.1178>

That the value of Lambda_QCD which is hardcoded.

Cheers,

Olivier

> On 27 Feb 2020, at 12:43, Sokratis Trifinopoulos <email address hidden> wrote:
>
> Question #688353 on MadGraph5_aMC@NLO changed:
> https://answers.launchpad.net/mg5amcnlo/+question/688353
>
> Sokratis Trifinopoulos posted a new comment:
> So it is always 3.5562e-06? In that case why is it 10^3 smaller than
> Lambda_QCD = 218MeV ?
>
> --
> You received this question notification because you are an answer
> contact for MadGraph5_aMC@NLO.